Update app.py

app.py

@@ -33,7 +33,13 @@ from qdrant_client.models import (
     FusionQuery,
     Fusion,
     SearchRequest,
-    Modifier
+    Modifier,
+    OptimizersConfigDiff,
+    HnswConfigDiff,
+    Distance,
+    VectorParams,
+    SparseVectorParams,
+    SparseIndexParams
 )
 
 def make_points(texts: List[str], metadatas: List[dict], dense: List[List[float]], sparse: List[SparseEmbedding])-> List[PointStruct]:
@@ -161,26 +167,26 @@ def load_models_and_documents():
     client.create_collection(
         collection_name,
         {
-            "text-dense": models.VectorParams(
+            "text-dense": VectorParams(
                 size=1024,
-                distance=models.Distance.COSINE,
+                distance=Distance.COSINE,
                 on_disk=False
             )
         },
         {
-            "text-sparse": models.SparseVectorParams(
-                index=models.SparseIndexParams(
+            "text-sparse": SparseVectorParams(
+                index=SparseIndexParams(
                     on_disk=False
                 ),
                 modifier=Modifier.IDF
             )
         },
         2,
-        optimizers_config=models.OptimizersConfigDiff(
+        optimizers_config=OptimizersConfigDiff(
             indexing_threshold=0,
             default_segment_number=4
         ),
-        hnsw_config=models.HnswConfigDiff(
+        hnsw_config=HnswConfigDiff(
             on_disk=False, 
             m=64,
             ef_construct=512
@@ -273,7 +279,7 @@ def load_models_and_documents():
         )
     client.update_collection(
         collection_name=collection_name,
-        optimizer_config=models.OptimizersConfigDiff(indexing_threshold=20000)
+        optimizer_config=OptimizersConfigDiff(indexing_threshold=20000)
     )       
 
     return client, collection_name, tokenizer, model, llm, dense_model, sparse_model
@@ -379,7 +385,7 @@ if __name__ == '__main__':
             else:
                 stop_token_ids = [151329, 151336, 151338]
                 sampling_params = SamplingParams(temperature=0.75, max_tokens=35, stop_token_ids=stop_token_ids)
-                prompt = [{"role": "user", "content": "hello"}]
+                prompt = [{"role": "user", "content": f"{}\nExplain the above in one sentence:"}]
                 inputs = tokenizer.apply_chat_template(prompt, tokenize=False, add_generation_prompt=True)
                 outputs = llm.generate(prompts=inputs, sampling_params=sampling_params)